Pancreatic cancer continues to present a major cause of cancer mortality in this country in that of the -25,000 new cases diagnosed each year, nearly all will succumb within a few months of detection. This bleak outlook is due to the absence of specific and sensitive diagnostic markers for the early stages of the disease and the lack of decisive means of tumor eradication at the later stages of progression and metastasis. The long range goal of this proposal is to utilize a battery of molecular probes that recognize normal human exocrine pancreatic cell constituents, several of which may be expressed specifically by subsets of pancreatic tumor cells of similar morphology. Poorly- differentiated and moderately-well differentiated pancreatic adenocarcinomas, for example, may be found to resemble several phases of normal tissue differentiation or a number of forms of transformed cells deriving from a particular normal cell type. Several lines of evidence indicate that pancreatic tumors can exhibit phenotypes that correspond to stages of the normal differentiation program. In these studies we plan to use riboprobes for several classes of mRNA: major cell structural components, enzymes involved in metabolism and secretion, hormone receptors, and protein kinases and substrates; antibodies specific for protein kinases and their substrates will be used in consort with gene probes for these molecules. Freshly fixed and snap-frozen pancreatic adenocarcinoma resections and biopsies and normal tissue will be analyzed using the techniques of Northern analysis of tissue extracts and in situ hybridization of tissue sections, as well as light microscopic immunocytochemistry for proteins of interest. Electron microscopy of sections of Epon- embedded fixed tissues will be done for refined morphological analysis. These specimens will be classified pathologically and the patients' clinical course correlated with the data. Sections of pathologic specimens embedded in paraffin from patients with pancreatic adenocarcinoma previously treated in our institution will be similarly examined in order to provide a retrospective clinical-pathologic correlation with new information gained in this study. At the same time, we plan to examine the normal embryonic rat pancreas with the same gene probes and immunologic probes in order to compare the human tumors with an accessible pancreatic developmental system. This analytical scheme should enable us to test the hypothesis that human pancreatic neoplasms reflect aberrant stages in the normal differentiation process and should allow us to identify potential cells of origin and developmental stages that may be the targets of the carcinogenic process in the pancreas.